SHORT TERM ALBANIAN GDP FORECAST: “ONE QUARTER TO ONE YEAR AHEAD”

In forecasting, macroeconomic variables such as GDP play an important role for policy makers and for the assessment of the future state of the economy. In this paper, different models to forecast quarterly GDP growth in Albania will be presented. The first group of models bases on ARIMA structures. These models are applied directly once on GDP series and then to the main economic activities which are been used to derive GDP. The second group of models for forecast uses VAR model, and the last group are refereed on by the bridge models. In cases were bridge models are used, variables with different frequencies are forecasted for the missing period. After that, all the series are aggregated at quarterly frequencies and are used for GDP forecast. Hence, the bulk of the material is to give comparisons of those models to forecast Albanian quarterly economic growth from one quarter up to four quarters ahead in a pseudo-real time setup